Recommendation method and device

Abstract

The invention discloses a recommendation method and device. The method comprises the steps: determining a bird nest with the best fitness based on a user or service scoring matrix; updating the bird nest by using Levy flight, and recording the optimal bird nest; performing crossover operation on the updated bird nest to obtain a new bird nest, and updating the optimal bird nest; updating the birdnest with the foreign bird egg found by the host, continuously updating the bird nest by using mutation operation, and updating the optimal bird nest; and outputting the optimal bird nest meeting thecondition as a service recommendation result of the user or a user recommendation result of the service. On the basis of an existing collaborative filtering algorithm, the advantages of high robustness and high convergence rate of a cuckoo algorithm are combined, a crossover operator of a genetic algorithm is fused into an improved cuckoo algorithm in a global search stage, and a mutation operatorof the genetic algorithm is fused into a local search stage; the advantages of high robustness and high convergence rate of the cuckoo algorithm are combined, the diversity and search capability of the population of the cuckoo algorithm are improved, and the quality of recommendation results is improved.

Description

technical field [0001] The invention belongs to the technical field of information recommendation, and in particular relates to a recommendation method and device. Background technique [0002] The current user / service recommendation methods mainly include: content-based recommendation, knowledge-based recommendation, data mining-based recommendation, network structure-based recommendation, collaborative filtering recommendation, etc. Among them, the collaborative filtering algorithm is widely used in e-commerce, big data and other fields because it can calculate the user's own behavior records and easily discover the user's potential information preference characteristics. [0003] In order to improve the quality of recommendation, some studies combine collaborative filtering algorithm and genetic algorithm to improve the recommendation effect. However, in the face of large amounts of data, the convergence speed of genetic algorithm becomes slower and the search ability is ...

AI Technical Summary

Problems solved by technology

[0003] In order to improve the quality of recommendation, some studies combine the collaborative filtering algorithm with the genetic algorithm to improve the recommendation effect. However, in the face of a large amount of data, the convergence speed of the genetic algorithm is slowed down and the search ability is weakened, which affects the real-time recommendation results. sex

Images